vllm的部署和使用

原创已于 2025-01-03 11:34:54 修改 · 954 阅读

0 ·

CC 4.0 BY-SA版权

文章标签：

#vllm

于 2024-12-31 17:41:52 首次发布

conda create -n cosyvoice python=3.10.9 cudnn=9.1.1.17 nvidia/label/cuda-12.1.1::cuda-toolkit ffmpeg x264

conda install -c conda-forge gcc_linux-64=12

pip install ninja

pip install torch==2.4.0 torchvision==0.19.0 torchaudio==2.4.0 xformers==0.0.27.post2 --index-url https://download.pytorch.org/whl/cu121

pip install vllm==0.6.6 transformers==4.46 modelscope==1.20.1

qwen2.5模型下载

from modelscope import snapshot_download

# Downloading model checkpoint to a local dir model_dir
# model_dir = snapshot_download('Qwen/Qwen2.5-0.5B-Instruct')
# model_dir = snapshot_download('Qwen/Qwen2.5-7B-Instruct')
# model_dir = snapshot_download('Qwen/Qwen2.5-32B-Instruct')
# model_dir = snapshot_download('Qwen/Qwen2.5-72B-Instruct')
model_dir = snapshot_download('Qwen/Qwen2.5-1.5B-Instruct')

vllm中文文档：https://vllm.hyper.ai/